Relapse patterns after chemo-radiation for carcinoma of the oesophagus.

AIM: The detailed review of patterns of failure in this report was undertaken to identify the continuing obstacles to the successful management of oesophageal cancer, and to establish whether there is a case to compare definitive chemo-radiation (Def-CR) and surgery for patients with squamous cancer in a randomized controlled trial. MATERIALS AND METHODS: First and subsequent sites of failure were reviewed in 274 patients treated with Def-CR using two cycles of cisplatin, infusional fluorouracil and 60 Gy; and 92 patients with limited chemo-radiation (CR), using one cycle and 35 Gy, followed by surgery (CR-Surg). All were treated on prospective non-randomized trials run by the Trans-Tasman Radiation Oncology Group between 1985 and 1999. Failure patterns were analysed using competing risks methodology, and pre-treatment variables predicting survival were identified by proportional hazards modelling. RESULTS: Site, stage, performance status and gender were independently predictive of survival following Def-CR. Local failure was evident in 42.3% of patients, but distant failure in isolation occurred in an additional 18.1%. Lowest rates of local and distant failure at 5 years (29.9% and 26%) occurred in patients with squamous cancer (SCC) located in the upper-third, whose 5-year survival was also the most favourable (49.2%). Survival was least favourable in patients with adenocarcinoma (AC) in the lower two-thirds (18.1%) due to higher rates of local (51.5%) and distant (36.1%) failure. Local failure occurred in 31.5% of patients undergoing CR-Surg but distant failure in isolation was observed in a further 34.7%. Outcomes were least favourable in patients with AC of the lower-third in whom 57.7% failed distantly and 5-year survival was 3.8%. Response to pre-operative chemo-radiation was also strongly predictive of outcome. Patients with no residual cancer in the resection specimen had the lowest rates of local (0%) and distant (16.7%) failure and the best survival (64.9%). Survival in patients with residual cancer in nodes, however, was extremely poor (3.5%) with distant failure occurring in 66.7%. CONCLUSION: The concurrent administration of chemotherapy with radiotherapy seems to have improved loco-regional control and has exposed distant failure as an obstacle to further improvements in outcome. Site, histological subtype, gender and response to chemo-radiation may predict biological differences in oesophageal cancer (OC) that influence outcome. A good case for a randomized comparison between Def-CR and CR-Surg in patients with SCC in the lower two-thirds exists.

Pilot study of high-frequency ultrasound to assess cutaneous oedema in the conservatively managed breast.

Cutaneous oedema is a relatively frequent complication in patients treated conservatively for breast cancer. The factors that contribute to this complication have not been precisely determined. We performed a pilot study to assess the usefulness of high-frequency ultrasound as a quantitative measure of cutaneous oedema. Eleven patients undergoing breast-conserving therapy for breast cancer were studied. Both the treated and untreated breasts were examined. Total cutaneous thickness provided a useful measure of cutaneous oedema. The treated breast was significantly thicker than the untreated breast (P < 0.001). The medial aspect of the breast was thicker than the lateral aspect in both the treated and untreated breast (P < 0.001). The increase in cutaneous thickness predated radiotherapy in those patients who had undergone an axillary dissection. Intrapatient variation in skin thickness was much less than interpatient variation in skin thickness (coefficient of variation 6.4% vs. 18.2% for the untreated breast; coefficient of variation 13.9% vs. 30.9% for the treated breast). Increasing cutaneous thickness was associated with decreasing cutaneous echodensity. We were unable to derive quantitative estimates of echodensity. Cutaneous oedema is an important outcome variable following conservative treatment of breast cancer. High-frequency ultrasound is able to quantify this accurately. It can readily detect changes invisible to the naked eye. High-frequency ultrasound should enable the effects of different treatment options (e.g., extent of surgery, radiotherapy, and chemotherapy) on cutaneous oedema to be differentiated and for the time course of oedema to be accurately characterised.